Package com.google.protobuf

Source Code of com.google.protobuf.JsonFormat$Tokenizer

// Protocol Buffers - Google's data interchange format
// Copyright 2008 Google Inc. All rights reserved.
// http://code.google.com/p/protobuf/
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

package com.google.protobuf;

import java.io.IOException;
import java.math.BigInteger;
import java.nio.CharBuffer;
import java.util.Iterator;
import java.util.List;
import java.util.Locale;
import java.util.Map;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

import com.google.protobuf.ByteString;
import com.google.protobuf.ExtensionRegistry;
import com.google.protobuf.Message;
import com.google.protobuf.UnknownFieldSet;
import com.google.protobuf.Descriptors.Descriptor;
import com.google.protobuf.Descriptors.EnumDescriptor;
import com.google.protobuf.Descriptors.EnumValueDescriptor;
import com.google.protobuf.Descriptors.FieldDescriptor;

/**
* Copied from http://code.google.com/p/protobuf/issues/detail?id=82
*
* Provide ascii text parsing and formatting support for proto2 instances. The implementation
* largely follows google/protobuf/text_format.cc.
*
* @author aantonov@orbitz.com Alex Antonov
*         <p/>
*         Based on the original code by:
* @author wenboz@google.com Wenbo Zhu
* @author kenton@google.com Kenton Varda
*/
public final class JsonFormat {

    /**
     * Outputs a textual representation of the Protocol Message supplied into the parameter output.
     * (This representation is the new version of the classic "ProtocolPrinter" output from the
     * original Protocol Buffer system)
     */
    public static void print(Message message, Appendable output) throws IOException {
        JsonGenerator generator = new JsonGenerator(output);
        generator.print("{");
        print(message, generator);
        generator.print("}");
    }

    /**
     * Outputs a textual representation of {@code fields} to {@code output}.
     */
    public static void print(UnknownFieldSet fields, Appendable output) throws IOException {
        JsonGenerator generator = new JsonGenerator(output);
        generator.print("{");
        printUnknownFields(fields, generator);
        generator.print("}");
    }

    /**
     * Like {@code print()}, but writes directly to a {@code String} and returns it.
     */
    public static String printToString(Message message) {
        try {
            StringBuilder text = new StringBuilder();
            print(message, text);
            return text.toString();
        } catch (IOException e) {
            throw new RuntimeException("Writing to a StringBuilder threw an IOException (should never "
                                                       + "happen).",
                                       e);
        }
    }

    /**
     * Like {@code print()}, but writes directly to a {@code String} and returns it.
     */
    public static String printToString(UnknownFieldSet fields) {
        try {
            StringBuilder text = new StringBuilder();
            print(fields, text);
            return text.toString();
        } catch (IOException e) {
            throw new RuntimeException("Writing to a StringBuilder threw an IOException (should never "
                                                       + "happen).",
                                       e);
        }
    }

    private static void print(Message message, JsonGenerator generator) throws IOException {

        for (Iterator<Map.Entry<FieldDescriptor, Object>> iter = message.getAllFields().entrySet().iterator(); iter.hasNext();) {
            Map.Entry<FieldDescriptor, Object> field = iter.next();
            printField(field.getKey(), field.getValue(), generator);
            if (iter.hasNext()) {
                generator.print(",");
            }
        }
        printUnknownFields(message.getUnknownFields(), generator);
    }

    public static void printField(FieldDescriptor field, Object value, JsonGenerator generator) throws IOException {

        printSingleField(field, value, generator);
    }

    private static void printSingleField(FieldDescriptor field,
                                         Object value,
                                         JsonGenerator generator) throws IOException {
        if (field.isExtension()) {
            generator.print("[");
            generator.print("\"");
            // We special-case MessageSet elements for compatibility with proto1.
            if (field.getContainingType().getOptions().getMessageSetWireFormat()
                && (field.getType() == FieldDescriptor.Type.MESSAGE) && (field.isOptional())
                // object equality
                && (field.getExtensionScope() == field.getMessageType())) {
                generator.print(field.getMessageType().getFullName());
            } else {
                generator.print(field.getFullName());
            }
            generator.print("\"");
            generator.print("]");
        } else {
            generator.print("\"");
            if (field.getType() == FieldDescriptor.Type.GROUP) {
                // Groups must be serialized with their original capitalization.
                generator.print(field.getMessageType().getName());
            } else {
                generator.print(field.getName());
            }
            generator.print("\"");
        }

        // Done with the name, on to the value

        if (field.getJavaType() == FieldDescriptor.JavaType.MESSAGE) {
            generator.print(": ");
            generator.indent();
        } else {
            generator.print(": ");
        }


        if (field.isRepeated()) {
            // Repeated field. Print each element.
            generator.print("[");
            for (Iterator iter = ((List) value).iterator(); iter.hasNext();) {
                printFieldValue(field, iter.next(), generator);
                if (iter.hasNext()) {
                    generator.print(",");
                }
            }
            generator.print("]");
        } else {
            printFieldValue(field, value, generator);
            if (field.getJavaType() == FieldDescriptor.JavaType.MESSAGE) {
                generator.outdent();
            }
        }
    }

    private static void printFieldValue(FieldDescriptor field, Object value, JsonGenerator generator) throws IOException {
        switch (field.getType()) {
            case INT32:
            case INT64:
            case SINT32:
            case SINT64:
            case SFIXED32:
            case SFIXED64:
            case FLOAT:
            case DOUBLE:
            case BOOL:
                // Good old toString() does what we want for these types.
                generator.print(value.toString());
                break;

            case UINT32:
            case FIXED32:
                generator.print(unsignedToString((Integer) value));
                break;

            case UINT64:
            case FIXED64:
                generator.print(unsignedToString((Long) value));
                break;

            case STRING:
                generator.print("\"");
                generator.print(escapeText((String) value));
                generator.print("\"");
                break;

            case BYTES: {
                generator.print("\"");
                generator.print(escapeBytes((ByteString) value));
                generator.print("\"");
                break;
            }

            case ENUM: {
                generator.print("\"");
                generator.print(((EnumValueDescriptor) value).getName());
                generator.print("\"");
                break;
            }

            case MESSAGE:
            case GROUP:
                generator.print("{");
                print((Message) value, generator);
                generator.print("}");
                break;
        }
    }

    private static void printUnknownFields(UnknownFieldSet unknownFields, JsonGenerator generator) throws IOException {
        for (Map.Entry<Integer, UnknownFieldSet.Field> entry : unknownFields.asMap().entrySet()) {
            String prefix = entry.getKey().toString() + ": ";
            UnknownFieldSet.Field field = entry.getValue();

            for (long value : field.getVarintList()) {
                generator.print("\"");
                generator.print(entry.getKey().toString());
                generator.print("\"");
                generator.print(": ");
                generator.print(unsignedToString(value));
                generator.print("\n");
            }
            for (int value : field.getFixed32List()) {
                generator.print("\"");
                generator.print(entry.getKey().toString());
                generator.print("\"");
                generator.print(": ");
                generator.print(String.format((Locale) null, "0x%08x", value));
                generator.print("\n");
            }
            for (long value : field.getFixed64List()) {
                generator.print(entry.getKey().toString());
                generator.print(": ");
                generator.print(String.format((Locale) null, "0x%016x", value));
                generator.print("\n");
            }
            for (ByteString value : field.getLengthDelimitedList()) {
                generator.print(entry.getKey().toString());
                generator.print(": \"");
                generator.print(escapeBytes(value));
                generator.print("\"\n");
            }
            for (UnknownFieldSet value : field.getGroupList()) {
                generator.print(entry.getKey().toString());
                generator.print(" {\n");
                generator.indent();
                printUnknownFields(value, generator);
                generator.outdent();
                generator.print("}\n");
            }
        }
    }

    /**
     * Convert an unsigned 32-bit integer to a string.
     */
    private static String unsignedToString(int value) {
        if (value >= 0) {
            return Integer.toString(value);
        } else {
            return Long.toString((value) & 0x00000000FFFFFFFFL);
        }
    }

    /**
     * Convert an unsigned 64-bit integer to a string.
     */
    private static String unsignedToString(long value) {
        if (value >= 0) {
            return Long.toString(value);
        } else {
            // Pull off the most-significant bit so that BigInteger doesn't think
            // the number is negative, then set it again using setBit().
            return BigInteger.valueOf(value & 0x7FFFFFFFFFFFFFFFL).setBit(63).toString();
        }
    }

    /**
     * An inner class for writing text to the output stream.
     */
    static private final class JsonGenerator {

        Appendable output;
        boolean atStartOfLine = true;
        StringBuilder indent = new StringBuilder();

        public JsonGenerator(Appendable output) {
            this.output = output;
        }

        /**
         * Indent text by two spaces. After calling Indent(), two spaces will be inserted at the
         * beginning of each line of text. Indent() may be called multiple times to produce deeper
         * indents.
         */
        public void indent() {
            indent.append("  ");
        }

        /**
         * Reduces the current indent level by two spaces, or crashes if the indent level is zero.
         */
        public void outdent() {
            int length = indent.length();
            if (length == 0) {
                throw new IllegalArgumentException(" Outdent() without matching Indent().");
            }
            indent.delete(length - 2, length);
        }

        /**
         * Print text to the output stream.
         */
        public void print(CharSequence text) throws IOException {
            int size = text.length();
            int pos = 0;

            for (int i = 0; i < size; i++) {
                if (text.charAt(i) == '\n') {
                    write(text.subSequence(pos, size), i - pos + 1);
                    pos = i + 1;
                    atStartOfLine = true;
                }
            }
            write(text.subSequence(pos, size), size - pos);
        }

        /**
         * Appends text to the output stream without starting a new line after it.
         */
        public void append(CharSequence text) throws IOException {
            int size = text.length();
            int pos = 0;

            for (int i = 0; i < size; i++) {
                if (text.charAt(i) == '\n') {
                    write(text.subSequence(pos, size), i - pos + 1);
                    pos = i + 1;
                    atStartOfLine = false;
                }
            }
            write(text.subSequence(pos, size), size - pos);
        }

        private void write(CharSequence data, int size) throws IOException {
            if (size == 0) {
                return;
            }
            if (atStartOfLine) {
                atStartOfLine = false;
                output.append(indent);
            }
            output.append(data);
        }
    }

    // =================================================================
    // Parsing

    /**
     * Represents a stream of tokens parsed from a {@code String}.
     * <p/>
     * <p>
     * The Java standard library provides many classes that you might think would be useful for
     * implementing this, but aren't. For example:
     * <p/>
     * <ul>
     * <li>{@code java.io.StreamTokenizer}: This almost does what we want -- or, at least, something
     * that would get us close to what we want -- except for one fatal flaw: It automatically
     * un-escapes strings using Java escape sequences, which do not include all the escape sequences
     * we need to support (e.g. '\x').
     * <li>{@code java.util.Scanner}: This seems like a great way at least to parse regular
     * expressions out of a stream (so we wouldn't have to load the entire input into a single
     * string before parsing). Sadly, {@code Scanner} requires that tokens be delimited with some
     * delimiter. Thus, although the text "foo:" should parse to two tokens ("foo" and ":"), {@code
     * Scanner} would recognize it only as a single token. Furthermore, {@code Scanner} provides no
     * way to inspect the contents of delimiters, making it impossible to keep track of line and
     * column numbers.
     * </ul>
     * <p/>
     * <p>
     * Luckily, Java's regular expression support does manage to be useful to us. (Barely: We need
     * {@code Matcher.usePattern()}, which is new in Java 1.5.) So, we can use that, at least.
     * Unfortunately, this implies that we need to have the entire input in one contiguous string.
     */
    private static final class Tokenizer {

        private final CharSequence text;
        private final Matcher matcher;
        private String currentToken;

        // The character index within this.text at which the current token begins.
        private int pos = 0;

        // The line and column numbers of the current token.
        private int line = 0;
        private int column = 0;

        // The line and column numbers of the previous token (allows throwing
        // errors *after* consuming).
        private int previousLine = 0;
        private int previousColumn = 0;

        private static Pattern WHITESPACE = Pattern.compile("(\\s|(#.*$))+", Pattern.MULTILINE);
        private static Pattern TOKEN = Pattern.compile("[a-zA-Z_][0-9a-zA-Z_+-]*|" + // an
                                                                                     // identifier
                                                       "[0-9+-][0-9a-zA-Z_.+-]*|" + // a number
                                                       "\"([^\"\n\\\\]|\\\\.)*(\"|\\\\?$)|" + // a
                                                                                              // double-quoted
                                                                                              // string
                                                       "\'([^\"\n\\\\]|\\\\.)*(\'|\\\\?$)", // a
                                                                                            // single-quoted
                                                                                            // string
                                                       Pattern.MULTILINE);

        private static Pattern DOUBLE_INFINITY = Pattern.compile("-?inf(inity)?",
                                                                 Pattern.CASE_INSENSITIVE);
        private static Pattern FLOAT_INFINITY = Pattern.compile("-?inf(inity)?f?",
                                                                Pattern.CASE_INSENSITIVE);
        private static Pattern FLOAT_NAN = Pattern.compile("nanf?", Pattern.CASE_INSENSITIVE);

        /**
         * Construct a tokenizer that parses tokens from the given text.
         */
        public Tokenizer(CharSequence text) {
            this.text = text;
            matcher = WHITESPACE.matcher(text);
            skipWhitespace();
            nextToken();
        }

        /**
         * Are we at the end of the input?
         */
        public boolean atEnd() {
            return currentToken.length() == 0;
        }

        /**
         * Advance to the next token.
         */
        public void nextToken() {
            previousLine = line;
            previousColumn = column;

            // Advance the line counter to the current position.
            while (pos < matcher.regionStart()) {
                if (text.charAt(pos) == '\n') {
                    ++line;
                    column = 0;
                } else {
                    ++column;
                }
                ++pos;
            }

            // Match the next token.
            if (matcher.regionStart() == matcher.regionEnd()) {
                // EOF
                currentToken = "";
            } else {
                matcher.usePattern(TOKEN);
                if (matcher.lookingAt()) {
                    currentToken = matcher.group();
                    matcher.region(matcher.end(), matcher.regionEnd());
                } else {
                    // Take one character.
                    currentToken = String.valueOf(text.charAt(pos));
                    matcher.region(pos + 1, matcher.regionEnd());
                }

                skipWhitespace();
            }
        }

        /**
         * Skip over any whitespace so that the matcher region starts at the next token.
         */
        private void skipWhitespace() {
            matcher.usePattern(WHITESPACE);
            if (matcher.lookingAt()) {
                matcher.region(matcher.end(), matcher.regionEnd());
            }
        }

        /**
         * If the next token exactly matches {@code token}, consume it and return {@code true}.
         * Otherwise, return {@code false} without doing anything.
         */
        public boolean tryConsume(String token) {
            if (currentToken.equals(token)) {
                nextToken();
                return true;
            } else {
                return false;
            }
        }

        /**
         * If the next token exactly matches {@code token}, consume it. Otherwise, throw a
         * {@link ParseException}.
         */
        public void consume(String token) throws ParseException {
            if (!tryConsume(token)) {
                throw parseException("Expected \"" + token + "\".");
            }
        }

        /**
         * Returns {@code true} if the next token is an integer, but does not consume it.
         */
        public boolean lookingAtInteger() {
            if (currentToken.length() == 0) {
                return false;
            }

            char c = currentToken.charAt(0);
            return (('0' <= c) && (c <= '9')) || (c == '-') || (c == '+');
        }

        /**
         * If the next token is an identifier, consume it and return its value. Otherwise, throw a
         * {@link ParseException}.
         */
        public String consumeIdentifier() throws ParseException {
            for (int i = 0; i < currentToken.length(); i++) {
                char c = currentToken.charAt(i);
                if ((('a' <= c) && (c <= 'z')) || (('A' <= c) && (c <= 'Z'))
                    || (('0' <= c) && (c <= '9')) || (c == '_') || (c == '.') || (c == '"')) {
                    // OK
                } else {
                    throw parseException("Expected identifier. -" + c);
                }
            }

            String result = currentToken;
            // Need to clean-up result to remove quotes of any kind
            result = result.replaceAll("\"|'", "");
            nextToken();
            return result;
        }

        /**
         * If the next token is a 32-bit signed integer, consume it and return its value. Otherwise,
         * throw a {@link ParseException}.
         */
        public int consumeInt32() throws ParseException {
            try {
                int result = parseInt32(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw integerParseException(e);
            }
        }

        /**
         * If the next token is a 32-bit unsigned integer, consume it and return its value.
         * Otherwise, throw a {@link ParseException}.
         */
        public int consumeUInt32() throws ParseException {
            try {
                int result = parseUInt32(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw integerParseException(e);
            }
        }

        /**
         * If the next token is a 64-bit signed integer, consume it and return its value. Otherwise,
         * throw a {@link ParseException}.
         */
        public long consumeInt64() throws ParseException {
            try {
                long result = parseInt64(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw integerParseException(e);
            }
        }

        /**
         * If the next token is a 64-bit unsigned integer, consume it and return its value.
         * Otherwise, throw a {@link ParseException}.
         */
        public long consumeUInt64() throws ParseException {
            try {
                long result = parseUInt64(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw integerParseException(e);
            }
        }

        /**
         * If the next token is a double, consume it and return its value. Otherwise, throw a
         * {@link ParseException}.
         */
        public double consumeDouble() throws ParseException {
            // We need to parse infinity and nan separately because
            // Double.parseDouble() does not accept "inf", "infinity", or "nan".
            if (DOUBLE_INFINITY.matcher(currentToken).matches()) {
                boolean negative = currentToken.startsWith("-");
                nextToken();
                return negative ? Double.NEGATIVE_INFINITY : Double.POSITIVE_INFINITY;
            }
            if (currentToken.equalsIgnoreCase("nan")) {
                nextToken();
                return Double.NaN;
            }
            try {
                double result = Double.parseDouble(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw floatParseException(e);
            }
        }

        /**
         * If the next token is a float, consume it and return its value. Otherwise, throw a
         * {@link ParseException}.
         */
        public float consumeFloat() throws ParseException {
            // We need to parse infinity and nan separately because
            // Float.parseFloat() does not accept "inf", "infinity", or "nan".
            if (FLOAT_INFINITY.matcher(currentToken).matches()) {
                boolean negative = currentToken.startsWith("-");
                nextToken();
                return negative ? Float.NEGATIVE_INFINITY : Float.POSITIVE_INFINITY;
            }
            if (FLOAT_NAN.matcher(currentToken).matches()) {
                nextToken();
                return Float.NaN;
            }
            try {
                float result = Float.parseFloat(currentToken);
                nextToken();
                return result;
            } catch (NumberFormatException e) {
                throw floatParseException(e);
            }
        }

        /**
         * If the next token is a boolean, consume it and return its value. Otherwise, throw a
         * {@link ParseException}.
         */
        public boolean consumeBoolean() throws ParseException {
            if (currentToken.equals("true")) {
                nextToken();
                return true;
            } else if (currentToken.equals("false")) {
                nextToken();
                return false;
            } else {
                throw parseException("Expected \"true\" or \"false\".");
            }
        }

        /**
         * If the next token is a string, consume it and return its (unescaped) value. Otherwise,
         * throw a {@link ParseException}.
         */
        public String consumeString() throws ParseException {
            return consumeByteString().toStringUtf8();
        }

        /**
         * If the next token is a string, consume it, unescape it as a
         * {@link com.google.protobuf.ByteString}, and return it. Otherwise, throw a
         * {@link ParseException}.
         */
        public ByteString consumeByteString() throws ParseException {
            char quote = currentToken.length() > 0 ? currentToken.charAt(0) : '\0';
            if ((quote != '\"') && (quote != '\'')) {
                throw parseException("Expected string.");
            }

            if ((currentToken.length() < 2)
                || (currentToken.charAt(currentToken.length() - 1) != quote)) {
                throw parseException("String missing ending quote.");
            }

            try {
                String escaped = currentToken.substring(1, currentToken.length() - 1);
                ByteString result = unescapeBytes(escaped);
                nextToken();
                return result;
            } catch (InvalidEscapeSequence e) {
                throw parseException(e.getMessage());
            }
        }

        /**
         * Returns a {@link ParseException} with the current line and column numbers in the
         * description, suitable for throwing.
         */
        public ParseException parseException(String description) {
            // Note: People generally prefer one-based line and column numbers.
            return new ParseException((line + 1) + ":" + (column + 1) + ": " + description);
        }

        /**
         * Returns a {@link ParseException} with the line and column numbers of the previous token
         * in the description, suitable for throwing.
         */
        public ParseException parseExceptionPreviousToken(String description) {
            // Note: People generally prefer one-based line and column numbers.
            return new ParseException((previousLine + 1) + ":" + (previousColumn + 1) + ": "
                                      + description);
        }

        /**
         * Constructs an appropriate {@link ParseException} for the given {@code
         * NumberFormatException} when trying to parse an integer.
         */
        private ParseException integerParseException(NumberFormatException e) {
            return parseException("Couldn't parse integer: " + e.getMessage());
        }

        /**
         * Constructs an appropriate {@link ParseException} for the given {@code
         * NumberFormatException} when trying to parse a float or double.
         */
        private ParseException floatParseException(NumberFormatException e) {
            return parseException("Couldn't parse number: " + e.getMessage());
        }
    }

    /**
     * Thrown when parsing an invalid text format message.
     */
    public static class ParseException extends IOException {

        public ParseException(String message) {
            super(message);
        }
    }

    /**
     * Parse a text-format message from {@code input} and merge the contents into {@code builder}.
     */
    public static void merge(Readable input, Message.Builder builder) throws ParseException,
                                                                     IOException {
        JsonFormat.merge(input, ExtensionRegistry.getEmptyRegistry(), builder);
    }

    /**
     * Parse a text-format message from {@code input} and merge the contents into {@code builder}.
     */
    public static void merge(CharSequence input, Message.Builder builder) throws ParseException {
        merge(input, ExtensionRegistry.getEmptyRegistry(), builder);
    }

    /**
     * Parse a text-format message from {@code input} and merge the contents into {@code builder}.
     * Extensions will be recognized if they are registered in {@code extensionRegistry}.
     */
    public static void merge(Readable input,
                             ExtensionRegistry extensionRegistry,
                             Message.Builder builder) throws ParseException, IOException {
        // Read the entire input to a String then parse that.

        // If StreamTokenizer were not quite so crippled, or if there were a kind
        // of Reader that could read in chunks that match some particular regex,
        // or if we wanted to write a custom Reader to tokenize our stream, then
        // we would not have to read to one big String. Alas, none of these is
        // the case. Oh well.

        merge(toStringBuilder(input), extensionRegistry, builder);
    }

    private static final int BUFFER_SIZE = 4096;

    // TODO(chrisn): See if working around java.io.Reader#read(CharBuffer)
    // overhead is worthwhile
    private static StringBuilder toStringBuilder(Readable input) throws IOException {
        StringBuilder text = new StringBuilder();
        CharBuffer buffer = CharBuffer.allocate(BUFFER_SIZE);
        while (true) {
            int n = input.read(buffer);
            if (n == -1) {
                break;
            }
            buffer.flip();
            text.append(buffer, 0, n);
        }
        return text;
    }

    /**
     * Parse a text-format message from {@code input} and merge the contents into {@code builder}.
     * Extensions will be recognized if they are registered in {@code extensionRegistry}.
     */
    public static void merge(CharSequence input,
                             ExtensionRegistry extensionRegistry,
                             Message.Builder builder) throws ParseException {
        Tokenizer tokenizer = new Tokenizer(input);

        // Based on the state machine @ http://json.org/

        tokenizer.consume("{"); // Needs to happen when the object starts.
        while (!tokenizer.tryConsume("}")) { // Continue till the object is done
            mergeField(tokenizer, extensionRegistry, builder);
        }
    }

    /**
     * Parse a single field from {@code tokenizer} and merge it into {@code builder}. If a ',' is
     * detected after the field ends, the next field will be parsed automatically
     */
    private static void mergeField(Tokenizer tokenizer,
                                   ExtensionRegistry extensionRegistry,
                                   Message.Builder builder) throws ParseException {
        FieldDescriptor field;
        Descriptor type = builder.getDescriptorForType();
        ExtensionRegistry.ExtensionInfo extension = null;

        if (tokenizer.tryConsume("[")) {
            // An extension.
            StringBuilder name = new StringBuilder(tokenizer.consumeIdentifier());
            while (tokenizer.tryConsume(".")) {
                name.append(".");
                name.append(tokenizer.consumeIdentifier());
            }

            extension = extensionRegistry.findExtensionByName(name.toString());

            if (extension == null) {
                throw tokenizer.parseExceptionPreviousToken("Extension \""
                                                            + name
                                                            + "\" not found in the ExtensionRegistry.");
            } else if (extension.descriptor.getContainingType() != type) {
                throw tokenizer.parseExceptionPreviousToken("Extension \"" + name
                                                            + "\" does not extend message type \""
                                                            + type.getFullName() + "\".");
            }

            tokenizer.consume("]");

            field = extension.descriptor;
        } else {
            String name = tokenizer.consumeIdentifier();
            field = type.findFieldByName(name);

            // Group names are expected to be capitalized as they appear in the
            // .proto file, which actually matches their type names, not their field
            // names.
            if (field == null) {
                // Explicitly specify US locale so that this code does not break when
                // executing in Turkey.
                String lowerName = name.toLowerCase(Locale.US);
                field = type.findFieldByName(lowerName);
                // If the case-insensitive match worked but the field is NOT a group,
                if ((field != null) && (field.getType() != FieldDescriptor.Type.GROUP)) {
                    field = null;
                }
            }
            // Again, special-case group names as described above.
            if ((field != null) && (field.getType() == FieldDescriptor.Type.GROUP)
                && !field.getMessageType().getName().equals(name)) {
                field = null;
            }

            if (field == null) {
                throw tokenizer.parseExceptionPreviousToken("Message type \"" + type.getFullName()
                                                            + "\" has no field named \"" + name
                                                            + "\".");
            }
        }

        tokenizer.consume(":");
        boolean array = tokenizer.tryConsume("[");

        if (array) {
            while (!tokenizer.tryConsume("]")) {
                handleValue(tokenizer, extensionRegistry, builder, field, extension);
                tokenizer.tryConsume(",");
            }
        } else {
            handleValue(tokenizer, extensionRegistry, builder, field, extension);
        }

        if (tokenizer.tryConsume(",")) {
            // Continue with the next field
            mergeField(tokenizer, extensionRegistry, builder);
        }
    }

    private static void handleValue(Tokenizer tokenizer,
                                    ExtensionRegistry extensionRegistry,
                                    Message.Builder builder,
                                    FieldDescriptor field,
                                    ExtensionRegistry.ExtensionInfo extension) throws ParseException {

        Object value = null;
        if (field.getJavaType() == FieldDescriptor.JavaType.MESSAGE) {
            value = handleObject(tokenizer, extensionRegistry, builder, field, extension);
        } else {
            value = handlePrimitive(tokenizer, field);
        }
        if (field.isRepeated()) {
            builder.addRepeatedField(field, value);
        } else {
            builder.setField(field, value);
        }
    }

    private static Object handlePrimitive(Tokenizer tokenizer, FieldDescriptor field) throws ParseException {
        Object value = null;
        switch (field.getType()) {
            case INT32:
            case SINT32:
            case SFIXED32:
                value = tokenizer.consumeInt32();
                break;

            case INT64:
            case SINT64:
            case SFIXED64:
                value = tokenizer.consumeInt64();
                break;

            case UINT32:
            case FIXED32:
                value = tokenizer.consumeUInt32();
                break;

            case UINT64:
            case FIXED64:
                value = tokenizer.consumeUInt64();
                break;

            case FLOAT:
                value = tokenizer.consumeFloat();
                break;

            case DOUBLE:
                value = tokenizer.consumeDouble();
                break;

            case BOOL:
                value = tokenizer.consumeBoolean();
                break;

            case STRING:
                value = tokenizer.consumeString();
                break;

            case BYTES:
                value = tokenizer.consumeByteString();
                break;

            case ENUM: {
                EnumDescriptor enumType = field.getEnumType();

                if (tokenizer.lookingAtInteger()) {
                    int number = tokenizer.consumeInt32();
                    value = enumType.findValueByNumber(number);
                    if (value == null) {
                        throw tokenizer.parseExceptionPreviousToken("Enum type \""
                                                                    + enumType.getFullName()
                                                                    + "\" has no value with number "
                                                                    + number + ".");
                    }
                } else {
                    String id = tokenizer.consumeIdentifier();
                    value = enumType.findValueByName(id);
                    if (value == null) {
                        throw tokenizer.parseExceptionPreviousToken("Enum type \""
                                                                    + enumType.getFullName()
                                                                    + "\" has no value named \""
                                                                    + id + "\".");
                    }
                }

                break;
            }

            case MESSAGE:
            case GROUP:
                throw new RuntimeException("Can't get here.");
        }
        return value;
    }

    private static Object handleObject(Tokenizer tokenizer,
                                       ExtensionRegistry extensionRegistry,
                                       Message.Builder builder,
                                       FieldDescriptor field,
                                       ExtensionRegistry.ExtensionInfo extension) throws ParseException {

        Object value;
        Message.Builder subBuilder;
        if (extension == null) {
            subBuilder = builder.newBuilderForField(field);
        } else {
            subBuilder = extension.defaultInstance.newBuilderForType();
        }

        tokenizer.consume("{");
        String endToken = "}";

        while (!tokenizer.tryConsume(endToken)) {
            if (tokenizer.atEnd()) {
                throw tokenizer.parseException("Expected \"" + endToken + "\".");
            }
            mergeField(tokenizer, extensionRegistry, subBuilder);
            if (tokenizer.tryConsume(",")) {
                // there are more fields in the object, so continue
                continue;
            }
        }

        value = subBuilder.build();
        return value;
    }

    // =================================================================
    // Utility functions
    //
    // Some of these methods are package-private because Descriptors.java uses
    // them.

    /**
     * Escapes bytes in the format used in protocol buffer text format, which is the same as the
     * format used for C string literals. All bytes that are not printable 7-bit ASCII characters
     * are escaped, as well as backslash, single-quote, and double-quote characters. Characters for
     * which no defined short-hand escape sequence is defined will be escaped using 3-digit octal
     * sequences.
     */
    static String escapeBytes(ByteString input) {
        StringBuilder builder = new StringBuilder(input.size());
        for (int i = 0; i < input.size(); i++) {
            byte b = input.byteAt(i);
            switch (b) {
                // Java does not recognize \a or \v, apparently.
                case 0x07:
                    builder.append("\\a");
                    break;
                case '\b':
                    builder.append("\\b");
                    break;
                case '\f':
                    builder.append("\\f");
                    break;
                case '\n':
                    builder.append("\\n");
                    break;
                case '\r':
                    builder.append("\\r");
                    break;
                case '\t':
                    builder.append("\\t");
                    break;
                case 0x0b:
                    builder.append("\\v");
                    break;
                case '\\':
                    builder.append("\\\\");
                    break;
                case '\'':
                    builder.append("\\\'");
                    break;
                case '"':
                    builder.append("\\\"");
                    break;
                default:
                    if (b >= 0x20) {
                        builder.append((char) b);
                    } else {
                        builder.append('\\');
                        builder.append((char) ('0' + ((b >>> 6) & 3)));
                        builder.append((char) ('0' + ((b >>> 3) & 7)));
                        builder.append((char) ('0' + (b & 7)));
                    }
                    break;
            }
        }
        return builder.toString();
    }

    /**
     * Un-escape a byte sequence as escaped using
     * {@link #escapeBytes(com.google.protobuf.ByteString)}. Two-digit hex escapes (starting with
     * "\x") are also recognized.
     */
    static ByteString unescapeBytes(CharSequence input) throws InvalidEscapeSequence {
        byte[] result = new byte[input.length()];
        int pos = 0;
        for (int i = 0; i < input.length(); i++) {
            char c = input.charAt(i);
            if (c == '\\') {
                if (i + 1 < input.length()) {
                    ++i;
                    c = input.charAt(i);
                    if (isOctal(c)) {
                        // Octal escape.
                        int code = digitValue(c);
                        if ((i + 1 < input.length()) && isOctal(input.charAt(i + 1))) {
                            ++i;
                            code = code * 8 + digitValue(input.charAt(i));
                        }
                        if ((i + 1 < input.length()) && isOctal(input.charAt(i + 1))) {
                            ++i;
                            code = code * 8 + digitValue(input.charAt(i));
                        }
                        result[pos++] = (byte) code;
                    } else {
                        switch (c) {
                            case 'a':
                                result[pos++] = 0x07;
                                break;
                            case 'b':
                                result[pos++] = '\b';
                                break;
                            case 'f':
                                result[pos++] = '\f';
                                break;
                            case 'n':
                                result[pos++] = '\n';
                                break;
                            case 'r':
                                result[pos++] = '\r';
                                break;
                            case 't':
                                result[pos++] = '\t';
                                break;
                            case 'v':
                                result[pos++] = 0x0b;
                                break;
                            case '\\':
                                result[pos++] = '\\';
                                break;
                            case '\'':
                                result[pos++] = '\'';
                                break;
                            case '"':
                                result[pos++] = '\"';
                                break;

                            case 'x':
                                // hex escape
                                int code = 0;
                                if ((i + 1 < input.length()) && isHex(input.charAt(i + 1))) {
                                    ++i;
                                    code = digitValue(input.charAt(i));
                                } else {
                                    throw new InvalidEscapeSequence("Invalid escape sequence: '\\x' with no digits");
                                }
                                if ((i + 1 < input.length()) && isHex(input.charAt(i + 1))) {
                                    ++i;
                                    code = code * 16 + digitValue(input.charAt(i));
                                }
                                result[pos++] = (byte) code;
                                break;

                            default:
                                throw new InvalidEscapeSequence("Invalid escape sequence: '\\" + c
                                                                + "'");
                        }
                    }
                } else {
                    throw new InvalidEscapeSequence("Invalid escape sequence: '\\' at end of string.");
                }
            } else {
                result[pos++] = (byte) c;
            }
        }

        return ByteString.copyFrom(result, 0, pos);
    }

    /**
     * Thrown by {@link JsonFormat#unescapeBytes} and {@link JsonFormat#unescapeText} when an
     * invalid escape sequence is seen.
     */
    static class InvalidEscapeSequence extends IOException {

        public InvalidEscapeSequence(String description) {
            super(description);
        }
    }

    /**
     * Like {@link #escapeBytes(com.google.protobuf.ByteString)}, but escapes a text string.
     * Non-ASCII characters are first encoded as UTF-8, then each byte is escaped individually as a
     * 3-digit octal escape. Yes, it's weird.
     */
    static String escapeText(String input) {
        return escapeBytes(ByteString.copyFromUtf8(input));
    }

    /**
     * Un-escape a text string as escaped using {@link #escapeText(String)}. Two-digit hex escapes
     * (starting with "\x") are also recognized.
     */
    static String unescapeText(String input) throws InvalidEscapeSequence {
        return unescapeBytes(input).toStringUtf8();
    }

    /**
     * Is this an octal digit?
     */
    private static boolean isOctal(char c) {
        return ('0' <= c) && (c <= '7');
    }

    /**
     * Is this a hex digit?
     */
    private static boolean isHex(char c) {
        return (('0' <= c) && (c <= '9')) || (('a' <= c) && (c <= 'f'))
               || (('A' <= c) && (c <= 'F'));
    }

    /**
     * Interpret a character as a digit (in any base up to 36) and return the numeric value. This is
     * like {@code Character.digit()} but we don't accept non-ASCII digits.
     */
    private static int digitValue(char c) {
        if (('0' <= c) && (c <= '9')) {
            return c - '0';
        } else if (('a' <= c) && (c <= 'z')) {
            return c - 'a' + 10;
        } else {
            return c - 'A' + 10;
        }
    }

    /**
     * Parse a 32-bit signed integer from the text. Unlike the Java standard {@code
     * Integer.parseInt()}, this function recognizes the prefixes "0x" and "0" to signify
     * hexidecimal and octal numbers, respectively.
     */
    static int parseInt32(String text) throws NumberFormatException {
        return (int) parseInteger(text, true, false);
    }

    /**
     * Parse a 32-bit unsigned integer from the text. Unlike the Java standard {@code
     * Integer.parseInt()}, this function recognizes the prefixes "0x" and "0" to signify
     * hexidecimal and octal numbers, respectively. The result is coerced to a (signed) {@code int}
     * when returned since Java has no unsigned integer type.
     */
    static int parseUInt32(String text) throws NumberFormatException {
        return (int) parseInteger(text, false, false);
    }

    /**
     * Parse a 64-bit signed integer from the text. Unlike the Java standard {@code
     * Integer.parseInt()}, this function recognizes the prefixes "0x" and "0" to signify
     * hexidecimal and octal numbers, respectively.
     */
    static long parseInt64(String text) throws NumberFormatException {
        return parseInteger(text, true, true);
    }

    /**
     * Parse a 64-bit unsigned integer from the text. Unlike the Java standard {@code
     * Integer.parseInt()}, this function recognizes the prefixes "0x" and "0" to signify
     * hexidecimal and octal numbers, respectively. The result is coerced to a (signed) {@code long}
     * when returned since Java has no unsigned long type.
     */
    static long parseUInt64(String text) throws NumberFormatException {
        return parseInteger(text, false, true);
    }

    private static long parseInteger(String text, boolean isSigned, boolean isLong) throws NumberFormatException {
        int pos = 0;

        boolean negative = false;
        if (text.startsWith("-", pos)) {
            if (!isSigned) {
                throw new NumberFormatException("Number must be positive: " + text);
            }
            ++pos;
            negative = true;
        }

        int radix = 10;
        if (text.startsWith("0x", pos)) {
            pos += 2;
            radix = 16;
        } else if (text.startsWith("0", pos)) {
            radix = 8;
        }

        String numberText = text.substring(pos);

        long result = 0;
        if (numberText.length() < 16) {
            // Can safely assume no overflow.
            result = Long.parseLong(numberText, radix);
            if (negative) {
                result = -result;
            }

            // Check bounds.
            // No need to check for 64-bit numbers since they'd have to be 16 chars
            // or longer to overflow.
            if (!isLong) {
                if (isSigned) {
                    if ((result > Integer.MAX_VALUE) || (result < Integer.MIN_VALUE)) {
                        throw new NumberFormatException("Number out of range for 32-bit signed integer: "
                                                        + text);
                    }
                } else {
                    if ((result >= (1L << 32)) || (result < 0)) {
                        throw new NumberFormatException("Number out of range for 32-bit unsigned integer: "
                                                        + text);
                    }
                }
            }
        } else {
            BigInteger bigValue = new BigInteger(numberText, radix);
            if (negative) {
                bigValue = bigValue.negate();
            }

            // Check bounds.
            if (!isLong) {
                if (isSigned) {
                    if (bigValue.bitLength() > 31) {
                        throw new NumberFormatException("Number out of range for 32-bit signed integer: "
                                                        + text);
                    }
                } else {
                    if (bigValue.bitLength() > 32) {
                        throw new NumberFormatException("Number out of range for 32-bit unsigned integer: "
                                                        + text);
                    }
                }
            } else {
                if (isSigned) {
                    if (bigValue.bitLength() > 63) {
                        throw new NumberFormatException("Number out of range for 64-bit signed integer: "
                                                        + text);
                    }
                } else {
                    if (bigValue.bitLength() > 64) {
                        throw new NumberFormatException("Number out of range for 64-bit unsigned integer: "
                                                        + text);
                    }
                }
            }

            result = bigValue.longValue();
        }

        return result;
    }
}
TOP

Related Classes of com.google.protobuf.JsonFormat$Tokenizer

TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.